Author round-table 019

InterviewsAuthor round-table

About adjustment of finished puzzles

The guests this time are Guten and Ogawa Minori. They often adjust their finished puzzles after they have sent them in. We asked them to tell about puzzle adjustments. From Nikoli we have NyanBaz and Nob.

NobI heard that you guys tend to make adjustments to your finished puzzles. I myself rarely adjust puzzles except when there is some defect. So I haven't thought a lot about adjustments. First, please tell us why you adjust your puzzles.

GutenI adjust my puzzles because they are not perfect at first, I'm not good at getting them perfect right away.

Ogawa MinoriI don't call a puzzle complete when I have composed it. It may have some other solution and I deal with that.

GutenIs it just checking the solution? Do you do a solution check and adjustments at the same time?

Ogawa MinoriYes. Sure, I adjust my puzzles when I check them out. Sometime it's done by just going over them 1 or 2 times. When it gets more than that, I repeat my solution checkup and I may do adjustments more than 5 times.

NobYou are in danger of introducing new other solutions when you adjust things. You need to do a solution checkup every time.

Ogawa MinoriSure, that is a given.

NobHow many times do you adjust a puzzle, Guten?

GutenI solve my puzzles a minimum of 5 or 6 times. And I may end up solving them more than 20 times. I don't go out of my way to adjust my puzzles all the time.

NobDo you do your adjustments to make it all more complete?

Ogawa MinoriYes, that's the case with me.

GutenI just said that I adjust puzzles because my work is often much less than perfect. But I don't do it to improve on the completeness. I adjust puzzles to get rid of poor parts. When I do a solution checkup, I check that a highlight hasn't faded away, I see that there isn't anywhere that leaves a poor impression, or something which may be disappointing. I want to make sure that's gone.

NobThen the result is a higher degree of completeness. What is something that is disappointing?

GutenFor example with a symmetric form in a numbers problem, a symmetric position that requires the same solution method. That's a very disappointing thing.

NobWell, I can understand that. If you adjust it, could there be something that becomes worse than in the original?

Ogawa MinoriThe very thing. If that happens I throw away the revision and return to the original.

GutenThat goes for me too.

NobHow about looking objectively at the problems you yourself made?

GutenThat's difficult. That's why I don't send in puzzles immediately after they are completed. I send in contributions at the earliest two months after they are done, and the wait may be up to a year. I always solve all puzzles before contributing them. There I adjust them again, it happens all the time.

NobSo you may solve a problem 20 times. And because you solve it so many times, you may learn the problem, like memorizing it?

GutenI do learn the problems that I solved a lot. I don't learn a problem that I solved just 5 or 6 times.

NobDo you make sure you solve your puzzles many times too, Minori?

Ogawa MinoriI just pass it on if I think it's complete after the solution checkup. I do that for all kinds of puzzles.

NyanBazI'm surprised to hear that. Isn't it difficult to adjust things with puzzles like Sudoku?

GutenWith Sudoku it is difficult. There are many cases where Sudoku doesn't hit the difficulty level that I aimed for. However, I let it be if it's fun. With Sudoku, when I replace numbers, rows, or columns, it seems like it becomes totally different problems.

Ogawa MinoriNumberlink changes when I replace numbers.

NyanBazAha, you guys adjust puzzles there too!

NobWhat's the kind of puzzle you find easy to adjust?

Ogawa MinoriOne puzzle which I adjust a lot is Nansuke. Replacing numbers doesn't change a Nansuke essentially. But to change the order in the lists matters. There are 18 numbers, and it has to become two columns. I divide them up with lines there.

NobYou are thinking of the space, the layout!

NyanBazPeople don't usually notice that kind of thing. Don't you worry about the things that are not noticed?

Ogawa MinoriNo problem there. (laughs)

GutenMinori's Heyawake problems leave me with a feeling of oddness, generally. I feel that you make them unnatural on purpose. When I adjust puzzles, I do it so they flow better.

Ogawa MinoriI'm very deliberate in my adjustments of Heyawake.

GutenIs that an adjustment of the mistakes or an adjustment of some place you don't like?

Ogawa MinoriBoth happen. Because my problem adjustment is conscious, it may become odd. I adjust Nurikabe a lot, the same with Heyawake.

GutenWhen I adjust Nurikabe, it becomes an easier problem. My problem changes give them a more simple and plain development.

Ogawa MinoriI like my problems to have a lot of white cells left with just 1 or 2 black ones. When I get an idea I revise what's around them.

NobThat's Minori all right.

GutenWith Nurikabe I go out of my way to see that difficult places become impossible if you work too hard on the solution.

Ogawa MinoriI adjust Nurikabe a lot, but Nurikabe is a puzzle where adjustment is difficult. It may end up in a totally different direction when I adjust to give it some specific direction. It's difficult to return to an early stage of the design with Nurikabe. Adjustment becomes very hard when I give the puzzles symmetrical placement especially where symmetry is not necessary. That's why I don't make Nurikabe or Heyawake with symmetric arrangements.

GutenI go for symmetric arrangements. It's a bother to adjust, but I don't mind that if it's fun. I feel lucky if I can adjust it and it is still a symmetric arrangement. When I give priority to how it should go and then adjust it, it may happen that the symmetric placement breaks down steadily and gets further from the original idea. But that is OK too. When I start to adjust, a problem becomes more refined and may become harder to solve. For example with Numberlink.

NobWhen you adjust puzzles, what do you pay attention to?

GutenI said that I adjust puzzles to get rid of weak points. But sometimes I think I may also want to follow up on a better idea. When I make a puzzle, I take pains to prepare for a highlight. If I think that one more highlight can be put somewhere, or there is a good introduction to reach this highlight, or if I think of something really wonderful, I may throw everything else out, highlight and all. Then it'll get shaken up and redone.

Ogawa MinoriI do the soultion checkup and adjustments to confirm the route to the solution. I'm careful to see that the route doesn't fork off and fade away.

GutenDo you try to make your problem solution so if follows a straightforward development?

Ogawa MinoriYes. I'm thinking about the scenario while making a puzzle. I want to fix the places where it veers off from what I aim at. So I may make many adjustments to delete what I think is a minus.

NobWhen you adjust a puzzle, can there be something that makes you change the whole scenario?

Ogawa MinoriNo, I don't change the scenario. I don't want to have to add something important later.

GutenI may change the story I set out to do. Even if it becomes a totally different problem, I think that's OK.

NyanBazWhat are the pluses and minuses in adjustments of puzzles?

Ogawa MinoriThe good point is that you are able to adjust story and difficulty so I can feel satisfied with it. The disadvantage is that it takes time.

GutenI think that adjustments are indispensable to avoid something disappointing, it is not something that is good or bad. But, I don't feel it becomes more complete even if I keep adjusting. I can get rid of bad points. But I only put a new highlight in maybe one of 7 or 8 problems. However, the final quality will improve. My point in adjustments are slightly different for different puzzles. When I explain all the details, it'll take forever. (laughs)

NobYou guys adjust puzzles to make better puzzles. What is a good puzzle then?

Ogawa MinoriIt's a puzzle that seems enjoyable when the solver has solved it. Even if it didn't go just as the author wanted, the story must make it go properly.

GutenSome kinds of puzzles have to be fun to be good. Some kinds of puzzles should have a few points that seem strange. I decide on that by the kind of puzzle. It may be important that it conveys what the author wants to express, but I don't mind it if that fails. I want you to enjoy my puzzles. For me, puzzles are entertainment not self-expression. I want to surprise solvers in something. That's a point where I want to add to it.

NobDo you cherish originality?

GutenWhat I go for is slightly different from originality. I hope to supplement what is insufficient in nikoli.com.

NobWell. There are parts where your thinking about adjustment are pretty much the same. And then there are some greatly different parts, too. Thinking about what you were saying we can maybe guess which parts were revised when we are solving your puzzles.


Discussion on Aug 2011 Published on Jun 18, 2014

Related link